Search

In recent years, power consumption has become one of the hottest research trends in computer science and industry. Most of the reasons are related to the operational budget and the environmental issues. In this paper, we would like to propose an energy-efficient solution for orchestrating the resource in cloud computing. In nature, the proposed approach firstly predicts the resource utilization of the upcoming period based on the Gaussian process regression method. Subsequently, the convex optimization technique is engaged to compute an appropriate quantity of physical servers for each monitoring window. This quantity of interest is calculated to ensure that a minimum number of servers can still provide an acceptable quality of service. Finally, a corresponding migrating instruction is issued to stack the virtual machines and turn off the idle physical servers to achieve the objective of energy savings. In order to evaluate the proposed method, we conduct the experiments using synthetic data from 29-day period of Google traces and real workload from the Montage open-source toolkit. Through the evaluation, we show that the proposed approach can achieve a significant result in reducing the energy consumption as well as maintaining the system performance. Balance between energy efficiency and quality of service in the cloud computing.Apply prediction technique to enhance the usefulness of monitoring statistics.Design optimal energy efficiency architecture to orchestrate the cloud system.After VM consolidation, adaptively turn-off idle physical machines to save energy.

Context information can be an important factor of user behavior modeling and various context recognition recommendations. However, state-of-the-art context modeling methods cannot deal with contexts of other dimensions such as those of users and items and cannot extract special semantics. On the other hand, some tasks for predicting multidimensional relationships can be used to recommend context recognition, but there is a problem with the generation recommendations based on a variety of context information. In this paper, we propose MRTensorCube, which is a large-scale data cube calculation based on distributed parallel computing using MapReduce computation framework and supports efficient context recognition. The basic idea of MRTensorCube is the reduction of continuous data combined partial filter and slice when calculating using a four-way algorithm. From the experimental results, it is clear that MRTensor is superior to all other algorithms.

As the vast amount of data in social Internet of Things (IoT) environments considering interactions between IoT and people is accumulated and processed through cloud and big data technologies, the services that utilize them are applied in various fields. The trust between IoT devices and their data is recognized as the core of IoT ecosystem creation and growth. Connection with suspicious IoT devices may pose a risk to services and system operation. Therefore, it is essential to analyze and manage trust information for devices, services, and people, as well as to provide the trust information to the other devices or users that need it. This paper presents a trust information management framework which contains a generic IoT reference model with trust capabilities to achieve the goal of converged trust information management. Additionally, a trust information management platform (TIMP) consisting of trust agents, trust information brokers, and trust information management systems has been proposed, which aims to provide trustworthy and safe interactions among people, virtual objects, and physical things. Implementing and deploying a TIMP enables a trustworthy ecosystem to be built while activating social IoT businesses by reducing transaction costs, as well as by eliminating the uncertainties in the use of social IoT services and data transactions.

Novel interactive activity recognition method using a topic modeling technique.Pose-based spatio-temporal relation features for intra- and inter-person.Two types of codeword corresponding to joint distance and angle.Four-level Pachinko Topic Model (PAM) for flexibly modeling interactions.Outperformance of recognition accuracy to state-of-the-art methods. In this paper, we go beyond the problem of recognizing video-based human interactive activities. We propose a novel approach that permits to deeply understand complex person-person activities based on the knowledge coming from human pose analysis. The joint coordinates of interactive objects are first located by an efficient human pose estimation algorithm. The relation features consisting of the intra and inter-person features of joint distance and angle, are suggested to use for describing the relationships between body components of the individual persons and the interacting two participants in the spatio-temporal dimension. These features are then provided to the codebook construction process, in which two types of codeword are generated corresponding to distance and angle features. In order to explain the relationships between poses, a flexible hierarchical topic model constructed by four layers is proposed using the Pachinko Allocation Model. The model is able to represent the full correlation between the relation features of body components as codewords, the interactive poselets as subtopics, and the interactive activities as super topics. Discrimination of complex activities presenting similar postures is further obtained by the proposed model. We validate our interaction recognition method on two practical data sets, the BIT-Interaction data set and the UT-Interaction data set. The experimental results demonstrate that the proposed approach outperforms recent interaction recognition approaches in terms of recognition accuracy.

The challenge that journalism is facing these days in the Internet mobile environment is greater than ever before. Journalism is losing its revenue structure to platform operators favoring a certain markets, and also the trust of its readers in light of fake news and infected news. To alleviate this situation, we propose a blockchain technology that is applicable to journalism in order to achieve decentralization as a reasonable alternative. The journalism model based on hybrid blockchain aims to achieve the following: the delivery of articles with sharing value, what we call proof of sharing, the distribution of roles of personalized agenda settings, and finally, the use of agora to collect public opinions. With all these, we attempt to resolve the issues with current journalism with our proposed model based on blockchain.

For the past ten years, Gaussian process has become increasingly popular for modeling numerous inferences and reasoning solutions due to the robustness and dynamic features. Particularly concerning regression and classification data, the combination of Gaussian process and Bayesian learning is considered to be one of the most appropriate supervised learning approaches in terms of accuracy and tractability. However, due to the high complexity in computation and data storage, Gaussian process performs poorly when processing large input dataset. Because of the limitation, this method is ill-equipped to deal with the large-scale system that requires reasonable precision and fast reaction rate. To improve the drawback, our research focuses on a comprehensive analysis of Gaussian process performance issues, highlighting ways to drastically reduce the complexity of hyper-parameter learning and training phases, which could be applicable in predicting the CPU utilization in the demonstrated application. In fact, the purpose of this application is to save the energy by distributively engaging the Gaussian process regression to monitor and predict the status of each computing node. Subsequently, a migration mechanism is applied to migrate the system-level processes between multi-core and turn off the idle one in order to reduce the power consumption while still maintaining the overall performance.

Recently, there are high demand on healthy life and wellness due to the increased lifespan. Wellness is generally used to meana healthy balance of the mind, body and spirit that results in an overall feeling of well-being. To monitor the physical and mental wellness of object, we developed an inspection service middlewarefor monitoring physical and mental health conditionbyanalyzing EEG (electroencephalography), ECG (electrocardiography), respiration rate, SpO2 and EMG (Electrocardiogram) waveforms from multi-modal biosensors under the coverage of a wireless sensor network (WSN).We also use ontologies are an adequate methodologyto model sensors and their capabilities. For the inspection service middleware, we propose a prediction model based onrisk ratioExpectation Maximization (EM) by monitoring bio-sensor data in real-time. We also used ontology model which enables reasoning, classification of datasets from sensor network and wellness contents recommendation. This inspection middleware for monitoringhealthcare condition and support recommendation of wellness contents such as customized exercise, proper diet, and hospital checkup. In this paper, there are the five modules as follows: (1) The measurement of biometrics such as body temperature,EEG, ECG, respiration rates and EMG, (2) Object assessment from measurement wavelength, (3) Situation assessment using GPS in smart device, (4) Maximized health condition using risk ratio EM, (5) decision making and recommendation of wellness contents.

People have multiple smart devices these days, such as smart phones, tablet computers and smart TVs among others. So now it is possible to offer advertisements everywhere. There is also an emerging technology with public Smart Digital Displays, which can be interacted with by smart mobile devices, by using the displays' touchscreen and other ways. Those displays offer a great chance for different companies to provide useful and interactive advertising for the customers. This is made possible by the Computer-Supported Cooperative Work and N-Screen technologies. And with the help of those technologies, it is possible to spread the advertisements to many more potential customers, because customers can share their experiences with their friends.

Smart devices are great tool to get a lot of information of user by variety sensors, application, web. The information is good clue to seize pattern of user. So, we can expect that customized content-service will be possible based on utilizing information of user. This expectation alters the type of content-service from just providing lots of contents to smart devices to recommendation contents which wanted, needed, favorite looking by user. For this customized content-service, a system model like a curator in galleries or museums is required. So, in this paper, we suggest Intelligent Contents Curation(ICCuration) model which has 3 sub modules with sensing, analysis and filtering information of user. The collected information of user are processed up to scenarios and the scenario is a clue for selecting contents which will be recommended to users. In the scenario has user`s preferences and behaviors as well as devices informations as elements. So, contents can be optimized not only domain category but type of media for devices.

The learners using learning content through the smart devices can access to the Internet from anytime and anywhere. However, with the rapid increase of learning content on the Web, it will be

Smart phone allows advertisers to estimate customers behavior by selecting user context awareness information and gives users instant feed back about their behavior. Electronic equipments such as smart phone enable advertisers to advertise interesting product for each customers at the point of purchase. In this paper, we deal with the trends of Smart phone and internet based TV in the spotlight as the upcoming advertising media and propose the effective way of advertising, Smart Advertising model, which can give users advertising contents of their interesting product by collecting user context information from a variety of devices including N-screen in smart space. This model will induce modern people who live in flood of advertisements to buy products by providing interesting advertising contents.

Context-aware technologies can make e-learning services smarter and more efficient since context-aware services are based on the user's behavior. To add those technologies into existing e-learning services, a service architecture model is needed to transform the existing e-learning environment, which is situation-aware, into the environment that understands context as well. The context-awareness in e-learning may include the awareness of user profile and terminal context. In this paper, we propose a new notion of service that provides context-awareness to smart learning content in a cloud computing environment. We suggest the elastic four smarts (E4S)--smart pull, smart prospect, smart content, and smart push--concept to the cloud services so smart learning services are possible. The E4S focuses on meeting the users' needs by collecting and analyzing users' behavior, prospecting future services, building corresponding contents, and delivering the contents through cloud computing environment. Users' behavior can be collected through mobile devices such as smart phones that have built-in sensors. As results, the proposed smart e-learning model in cloud computing environment provides personalized and customized learning services to its users.

Recently, Internet users have an interest about Social Media Service in Web2.0 environment. We suggest the orchestra service as social media service to meet user satisfactions in changed web environment. We accept a concept of the MMMD (Multiple Media Multiple Devices). In other words, Internet users listen to the music not only one device but multiple devices. Each one of multiple devices can play a sound source under earmark instruments for providing users with actual feeling like an orchestra. To meet the purpose, we define 3 steps. First, we separate the sound source based on instrument information. Second, we exact the suitable sound source for play orchestra. In final step, the sound source transmits to each suitable playing device. We named the 3 step for AET process. Beside we suggest synchronization method using rest point in the MIDI file for control sound sources. Using the AET process and synchronization method we provide the orchestra service for meet user`s satisfactions to users.

In this paper, we go beyond the problem of recognizing human interactions using videos collected from CCTV-based surveillance systems. We propose an approach that permits to deeply describe common person-person activities in the daily life based on the human poses. The joint coordinates of detected human objects are first located by an impressive articulated-body estimation algorithm using the tree graphical structure technique. The relational features consisting of the intra and inter-person feature describing the joint distance and angle information are used for describing the relationships between body components of the individual persons and the interaction of two participants. Moreover, the interaction is also considered in the spatio-temporal dimension in order to upgrade the discrimination among complex activities having much homothetic representation. We validate our interaction recognition method on two practical datasets, the BIT-Interaction dataset and the UT-Interaction dataset, using the multi-class Support Vector Machine technique. The experimental results demonstrate that the proposed approach using pose-body features outperforms recent interaction recognition approaches in the term of classification accuracy.

Contextual information has been recognized as an important factor to consider in user-aware Recommendation Systems. Since contextual information can be used as a significant factor in modeling user behavior, various context-aware recommendation methods are proposed. However, the state-of-the-art context modeling methods treat contexts as other dimensions similar to the dimensions of users and items, and cannot extract the special semantic operation of contexts. On the other hand, some works on multi-domain relation prediction can be used for the context-aware recommendation, but they have problems in generating recommendation under a large amount of contextual information. In this paper, we propose the 4-way Tensor, a parallel tensor factorization algorithm, to accelerate the tensor factorization of large datasets to support efficient context-aware recommendations. The basic idea of this algorithm is to partition a tensor into partition and then exploit the inherent parallelism to perform tensor related operations in parallel.

Recently CCTV-based behavior recognition have gained considerable attention in the transportation surveillance systems to identify normalities, such as traffic jams, accidents, and dangerous driving. An improved method is presented in this paper for the traffic behavior surveillance system by discovering more highly specific features based on the trajectory information. The multiple sparse feature comprising the object location, moving direction, speed, and appearance time length obtained from the moving object detection and tracking stage is modeled by the Pachinko Allocation Model. This hierarchical probabilistic model captures the correlation among the traffic activities and behaviors through the sparse features as the visual words. In the classification phase, the Support Vector Machine constructed from Decision Tree Architecture is utilized. Compared with existing methods, the proposed method outperforms 3-8% approximately in overall classification accuracy.

In the recently, the Closed Circuit Television (CCTV) has been used to ensure the security and evidence for the crimes. However, the video captured from CCTV has being used in the postprocessing to apply to the evidence. The using pattern of CCTV shows a slight effect on the purpose of prevention a crime rather than prevention a pre-crime that occurs in practical situations. In this paper, we propose a Flexible Multi-Level model for estimating whether dangerous behavior risk by analyzing the behavior of the object using the data of the CCTV collected by pedestrian. The FML model consists of the three steps as follows; object filtering, situation analysis, and abnormal decision. The object filtering checks the environment and context for pedestrians. The situation analysis builds the knowledge for the pedestrians tracking. Finally, the decision step decides and notifies the threat situation when the behavior observed object is determined to abnormal behavior. It is possible to respond quickly before crime, which enables high-speed situations judgment.

Interactive activity recognition from the RGB videos still remains a challenge, therefore some existing approaches paid the attention to RGB-Depth video process to avoid problems relating to mutual occlusion and redundant human pose and to improve accuracy of skeleton extraction. From the single action to complex interaction activity, it is necessary an efficient model to describe the relationship of body components between multi-human objects. In this research, the authors proposed a hierarchical model based on the Pachinko Allocation Model for interaction recognition. Concretely, the joint features comprising joint distant and joint motion are calculated from the skeleton position and then support to topic modeling. The probabilistic models describing the flexible relationship between features — poselets — activities are generated by this model. Finally, the Binary Tree of Support Vector Machine is applied for classification. Compared with existing state-of-the-arts, the proposed method outperforms in overall classification accuracy (8–21% approximately) with the SBU Kinect Interaction Dataset.

CCTV-based behavior recognition systems have gained considerable attention in recent years in the transportation surveillance domain for identifying unusual patterns, such as traffic jams, accidents, dangerous driving and other abnormal behaviors. In this paper, a novel approach for traffic behavior modeling is presented for video-based road surveillance. The proposed system combines the pachinko allocation model (PAM) and support vector machine (SVM) for a hierarchical representation and identification of traffic behavior. A background subtraction technique using Gaussian mixture models (GMMs) and an object tracking mechanism based on Kalman filters are utilized to firstly construct the object trajectories. Then, the sparse features comprising the locations and directions of the moving objects are modeled by PAM into traffic topics, namely activities and behaviors. As a key innovation, PAM captures not only the correlation among the activities, but also among the behaviors based on the arbitrary directed acyclic graph (DAG). The SVM classifier is then utilized on top to train and recognize the traffic activity and behavior. The proposed model shows more flexibility and greater expressive power than the commonly-used latent Dirichlet allocation (LDA) approach, leading to a higher recognition accuracy in the behavior classification.

Emerging technologies, such as multimedia and mobile computing, require that middleware platforms can support high adaptability in order to provide more flexible services in heterogeneous environments. To support high quality of service in distributed wired/wireless environments, it will be necessary for applications and middleware to be aware of changes in users requirements as well as environmental conditions, also to be able to adapt their behaviour as such changes. According to the needs of adaptability supporting in middleware, we structure a component-based, mobile agents framework in base level and meta level by using reflection. We propose concepts of meta agents and meta-service agents that are able to reflect changes made by mobile agents to the system. At the meta level of our framework, meta agents monitor execution of mobile agents and it is possible to provide dynamic adaptation of users requirements, deployment of application services and service customization with meta agents and meta-service agents.

This paper proposes a UQoSM (User Qualify of Service Middleware) framework for multimedia application systems. UQoSM system is extended the existing event service model added to the event monitoring, reflective event filtering and event dispatcher for supporting multimedia application systems. Especially, this paper is concentrated on providing suitable reflective event filtering function for multimedia application service system in order to meet various user requirements under inter-based environment. It means this system provides high QoS to users. In addition, it results in decreasing network traffic as unnecessary event information is filtered from network.

The Closed Circuit Television (CCTV) systems have been used at large scale for tracking and getting popularity with every passing day. The most common goal of CCTV system is prevention of crime and disorder by tracking the objects. The smartphone world is also expanding at a rapid pace since the technology was first introduced. Most users of smartphones live in those countries where usage of CCTV system is part of daily life. This paper studies a tracking system for mobile users based on CCTV system, where information can be sent from mobile to server so that CCTV system can work more specifically and accurately by tracking and recognizing objects. A safety insurance approach is proposed, in which a user can inform his/her location for close observation via CCTV system if he/she feels a potential threat. In case of emergency, location, nature of problem and possible difficulties can be determined in comparatively less time by authorities as they have already monitoring the situation.

A novel approach for human behavior modelling is represented in this paper based on the Pachinko Allocation Model (PAM) algorithm for the video-based road surveillance. In particular, the authors focus on the behavior analysis and modelling for learning and training as the main distribution of this research. Sparse object features in sequence of frames are modelled into activities and behaviors with full topic correlations to avoid omissions of small activities.

Applications of private matching (PM) are not limited to any specific domain of interest. Social media, e-health and commerce applications can afford to share certain common information without disclosing any extra details. PM ensures to reveal only common information between the communicating parties. In this paper we have presented a light weight protocol named Reflection for PM. Depending upon different requirements of involving parties, Reflection can be used in presence of third party or as a point to point communication. Using services of third party (cloud) where communication has to route through it, it still preserves the privacy aspect of values being matched. The protocol works on a random number and produces the result of matching on behalf of original values, hence it converges in only two pass communication. It also protects any additional slipway of information during or after the PM. It safeguards the user communication against eavesdropper and even from offline attacks. In its simpler version, the protocol can work without using any encryption. Value comparison mechanism in presence of encryption, makes the overall process of PM more secure against hostile users.

One of the interesting and realistic application areas for context-aware and cloud computing technologies are the learning systems. In order to add those technologies to the e-learning services that already exist, there is a need for a service architecture model that transforms the already existing situation-aware e-learning environment to an environment that understands context also. We propose using elastic computing in cloud computing for a new notion in smart learning. Our suggestion is the elastic conductor which including the four smarts concept to the cloud services in order to make smart learning services possible. The benefits of smart computing for e-learning solution, which use the handle intelligently to meet the users' needs, are the focus of this paper. This is achieved by going through the collecting user's behavior, prospecting, building, delivering and rendering steps. The smart-learning model that is proposed in this paper will show that it is possible to make customized and personalized learning services in different fields.

Advances in technology for multimedia services have led to a tremendous growth of video contents and accelerated the need to analyze and understand video content. An analysis of sports video, for example, has been a hot research area to identify a number of potential items: player and ball. For the efforts, this paper shows a Cognitive TV framework for the semantic region of interests in sport video service. The framework will issue two contributions in the short classification and object trajectory for the area of sport video. For the contributions, this paper suggests the semantic region of interests (SROI) based on Motion Vector Space (MVF) to analyze sports video. The SROI distinguishes the shot classes that are located in the motion vector space and detects the key objects, like the player and ball in the playing field, using Cognitive lattice algorithm.

To ensure the correctness of a parallel transaction processing system, we should guarantee the transaction atomicity despite of failures such as transaction abort or processor crash. When a real-time system employs parallel processing, it must achieve failure atomicity with timeliness because its major goal is to meet various timing requirements. A variety of protocols for ensuring correctness in real-time processing have been proposed. However, these protocols did not consider both timeliness and correctness together for the parallel processing. They considered only one of two; timeliness or correctness. Because real-time transactions must guarantee the timeliness and the correct completion in the presence of failures, the parallel transaction processing in real-time systems must consider both timeliness and correctness. This paper proposes a new commit protocol, which is called Integrated Commitment Protocol (ICP). This protocol is composed of two parts: the one is language constructs to reflect real-time requirements and the other is the commit procedures to guarantee the failure atomicity.

The rapid advances in multimedia and e-learning technologies let us present text, music, video and oral instruction on computers in more vivid ways. The problem is that learning environment has just single information media. That does not include multimedia data and does not give interesting for user. This paper introduces a new e-learning model which supports the harmonized media based on technologies to create and synchronize for multimedia presentation. One of the fundamental concepts of technologies is a recommendation engine based on semantics analysis. For the harmonized media, MPEG-21 Digital Item Declaration is applied to the modeling of synchronized presentation with multi information media.

The cloud computing technology causes much attention in IT field. The developments using this technology have done actively. The cloud computing is more evolved than the existing offer. So, the current cloud computing only has a process that responds user requirements when users demand their needs. For intelligently adapting the needs, this paper suggests a intelligent smart cloud model that is based on 4S/3R. This model can handle intelligently to meet users needs through collecting user’s behaviors, prospecting, building, delivering, and rendering steps. It is because users have always mobile devices including smart phones so that is collecting user’s behavior by sensors mounted on the devices. The proposed service model using intelligent smart cloud computing will show the personalized and customized services to be possible in various fields.

The multimedia service delivery chain poses today many challenges. There are an increasing terminal diversity, network heterogeneity and a pressure to satisfy the user preferences. The situation encourages the need for the personalized contents to provide the user in the best possible experience in ubiquitous computing. This paper introduces a personalized content preparation and delivery framework for multimedia service. The personalized video adaptation is expected to satisfy individual users’ need in video content. Cognitive lattice plays a significant role of video annotation to meet users’ preference on video content. In this paper, a comprehensive solution for the PVA (Personalized Video Adaptation) is proposed based on Cognitive lattice concept. The PVA is implemented based on MPEG-21 Digital Item Adaptation framework. One of the challenges is how to quantify users’ preference on video content.

Video adaptation has been introduction to tailor the same video to different derived user contexts to allow end users with different access network, client device and user profiles to access the same information source. Therefore, there is a need for the increase of automated auditing and intelligent reporting mechanisms for the download digital contents. This paper presents a proposed architecture of an Intelligent Adaptation Model (IAM). The objective of this system investigation into the potential security issues involved in content adaptation, especially video content adaptation, in multimedia intelligent system. For delivery video content we suggest the intelligent adaptation model that uses the digital item description in MPEG-21 multimedia framework.

In recent, service providers would like to support actual feeling services for internet users to give a satisfaction. Since web environments are changed more newly, internet users want to satisfy their needs. For offering the satisfaction to users, we suggest a middleware to give impressive media that is named Harmonized Media Service Middleware (HMSM). HMSM can support the harmonized media by using some emotional knowledge for each user. The harmonized media is constructed with media content, like videos, and emotional effects. The content is made by individually user, so it has a just recorded data as simple media. However, because user wants to a feeling of satisfaction, it is necessary to offer multiple media which include emotional effects for user's impression. In this paper, we suggest a way to offer a middleware for actual feeling multiple media, called HMSM. HMSM analyzes the input media and then extracts the effects which have compatibility with the original media from Media database. At the result, HMSM will be offer actual feeling service to satisfy user's needs.

Recently for media content from Internet users, the increased interest in the web is huge and rich. Between media content, especially in the eyes and ears of Internet users attracted to the limelight of the media content is video content. Additionally, in conjunction with Web 2.0 Internet users by their openness to each other and share content to create new forms of distribution of the flow. Video media content for distribution on the Web was created by a professional content to experts, but Web 2.0 era, the UGC (User Generated Content) in the form of self-produced content is the most. Made by the general Internet users, but self-produced content to provide video information only and has limitations. To satisfy Internet users as consumers, not just the images you add various effects are realized the need to provide the video. Therefore, this paper used ontology concept for providing media that has actual feeling instead of simple information media. Obtained Knowledge of input media by using ontology concept stand in the basis of extracting actual feeling effects. On the basis of this methodology, this paper proposes Fusion Media Middleware (FMM) that has function of offering fused media (music, sounds, images etc.). we describe the formatting guidelines for ACM SIG Proceedings.

The video clips only have simple information that these are just recorded data. There is a need to make the video clips actual sensation and look real to satisfy users for resolve the needs. We suggest new video model, called AFVM (Actual Feeling Video Model), that is based on ontology concept. The purpose of this study is to provide AFVM and new format of space where containing divided videos for actual image expression.

A ubiquitous learning environment provides an interoperable, pervasive and seamless learning architecture to connect, integrate, and share three major dimensions of learning resources: learning collaborators, learning contents, and learning services. Ubiquitous learning is characterized by providing intuitive ways for identifying right learning collaborators, right learning contents and right learning service in the any place at the any time. In this paper we propose of the two systems, collaborative adaptation system and personalization learning system. Personalized support for learners even more important, when E-learning takes place in open and dynamic learning and information network. Sometimes, users might prefer “emotional decision” to select their interested learning contents. The situation encourages the need for the personalized contents in order to provide the user in the best possible experience. This paper shows how to realize adaptation and personalized learning in e- Learning environments based on Cognitive lattice technique and Collaborative Adaptation Middleware.

Personalized video adaptation is expected to satisfy individual users' needs on video content. Multimedia data mining plays a significant role of video annotation to meet users' preference on video content. In this paper, a comprehensive solution for personalized video adaptation is proposed based on video content mining. Video content mining targets both cognitive content and affective content. Sometimes, users might prefer "emotional decision" to select their interested video content. The situation encourages the need for the personalized contents to provide the user in the best possible experience. We address the problem of video personalization. For the personalized content, we suggest the UVA (universal-video adaptation) model that uses the video content description in MPEG-7 standard and MPEG-21 multimedia framework.

To generate dynamically a personalized video summary, a video customization system is designed and implemented incorporating usage environment. The customization system uses the three tier architecture to adapt and deliver some media contents for the user. The multimedia contents delivery includes today many challenges. There are increasing the diversity of user devices and network interface devices, and the pressure to satisfy the user preferences. These situations encourage the need of the customized contents for the user. In this paper, we describe the customization of multimedia contents, like video. For the customization of content, we propose the Hybrid Multimedia Access (HMA) model that uses the multimedia content descriptions such as MPEG-7 standard and MPEG-21 multimedia framework.

As needs arise for an adaptive service with a new concept according to 3D graphic contents restrictions in the user environments, this study proposes an adaptive service model for 3D graphic contents in ubiquitous environment. In advance, we proposed Service Rendering Middleware to reconstruct the elements of existing contents according to the changing user environments as a middle layer model. This study provides a reference algorithm that selects an appropriate mesh model in Service Rendering Middleware. The algorithm provides a service level adequate for the current environment, allowing accurate creation and management of new 3D graphic contents according to device capabilities. An increasing amount of 3D graphic contents is being serviced to a wide range of terminal devices, and the approaches suggested in this paper are expected to enable fast adaptive services that enhance user satisfaction.

We present a filtered retrieval technique for structural information on internet-scale knowledge such as large-scale XML data in the Web. The technique evaluates XML standard queries on heterogeneous XML documents using information retrieval technique based on the relational tables in relational database management systems. The XML standard queries, XPath queries, in their general form are partial match queries, and these queries are particularly useful for searching documents of heterogeneous schemas. Thus, our technique is geared for partial match queries expressed as the queries. This indexes the elements in label paths, which are sequences of node labels, like keywords in texts, and finds the label paths matching a given query.

In this research we present a method for adaptation of 3D contents over ubiquitous environment. The need of contents adaptation and delivery is growing to support multiple target platforms for single source. Since contents base on computer graphics deal with a large data set and a high performance, a service adaptation for context changing is required to manipulate graphics contents with a more complicated method in multiple devices such as desktops, laptops, PDAs, mobile phones, etc. This paper presents some theoretical method for adaptation, which is service rendering middleware, and device-based decision making algorithm for LOD of 3D contents. Service rendering middleware consists of service adaptation (SA) for analyzing environments, service rendering (SR) for reconfiguring customized services and LOD algorithm for generating an adapted service by processing customized data according to the level of detail. Also, the decision-making algorithm, which is based on device capabilities, is used both in LOD algorithm module and method to decide level of 3D contents.

Character recognition has been an active research area in the field of pattern recognition. The existing character recognition algorithms are focused mainly on increasing the recognition rate. However, as in the recent Google Library Project, the requirement for speeding up recognition of enormous amount of documents is growing. Moreover, the existing algorithms do not pay enough attention to Asian characters. In this paper, we propose an algorithm for fast recognition of Asian characters based on the database methodologies. Since the number of Asian characters is very large and their shapes are complicated, Asian characters require much more recognition time than numeric and Roman characters. The proposed algorithm extracts the feature from each of Asian characters through the Discrete Fourier Transform (DFT) and optimizes the recognition speed by storing and retrieving the features using a multidimensional index. We improve the recognition speed of the proposed algorithm using the association rule technique, which is a widely adopted data mining technique. The proposed algorithm has the advantage that it can be applied regardless of the language, size, and font of the characters to be recognized.

As a result of the slow deployment of IPv6, many new proposals have been presented for the next generation Internet or to expand the IPv4 Internet. But none of them except NAT has been deployed successfully, mainly due to the requirement of modification of hosts and/or routers and the lack of scalable accessibility from the public IPv4 Internet to their realms. Recently UNA was presented to provide that scalable accessibility, with no modification of routers and hosts. It is still anticipated, however, that it would take a long time until IPv6 or other technology replaces IPv4 entirely. But it is urgent that the address space shortage problem be solved and that value-added services such as QoS and mobility be provided. In this paper, we discuss about the reorganization of the IPv4 Internet into a multi-tier architecture as a possible way of evolving the Internet, which can support a larger address space and deal with security, QoS, multicast, mobility and so on. As such an architecture, we present MTI (Multi-Tier Internet), in which UNA and MTAII are integrated, without forcing any modification of routers and hosts for easy deployment.

A high adaptable middleware has been an essential platform to provide more flexible services for multimedia, mobile, and ubiquitous applications in ubiquitous computing environments. In addition, the persistent services of these application systems and their middleware in ubiquitous computing are required so that they can be aware of the frequent and unpredictable changes in users’ requirements as well as environmental conditions and adapt their behavioural changes. However, current approaches for supporting adaptability have made applications themselves trigger and execute an adaptive mechanism when the underling infrastructure notifies them about any changes. In this paper, we propose a novel component-based context-adaptive model for context-awareness middleware to support efficiently dynamic adaptation of application services. We also demonstrate the current implementation of the context-adaptive middleware that help applications to adapt their ubiquitous computing environments according to rapidly changing contexts such as user-specific preferences, application-specific preferences, and low-level configurations.

An important requirement in geographic information systems is the ability to display multi-scale data. Existing spatial access methods do not access multi-scale data efficiently and do not deal with all types of multi-scale data. To resolve this, an efficient spatial access method in multi-scale GIS is described. A performance evaluation shows that our method outperforms existing spatial access methods.

Although various types of path indexes-indexes on path expressions-have been proposed for efficient processing of object-oriented queries, conventional join algorithms do not effectively utilize them. We propose a new join algorithm called OID join algorithm that effectively utilizes (multiple) path indexes in object-oriented databases. When (multiple) path indexes are available for a query, the OID join algorithm may reduce the query evaluation cost significantly by taking full advantage of the path indexes. We present a cost analysis for the OID join algorithm and compare it with those of conventional ones.

This shows a host DBMS simulator that can simulate directly switching software having Embedded SQL on a host computer, not switching systems. Execution files compiled on a host are loaded on the switching system, after testing on the switching testbed hardware. The testing on the testbed has difficult problems that cannot share more than one user simultaneously and can not test the relationship between two execution files. To solve the restriction, the host DBMS simulator has been developed. This is a simulator that can test execution files on the host without loading to the testbeds or switching systems, and can control simultaneously many query requests generated by multiple users on the same host. In this paper, we propose a DBMS simulator on the host system, and present in detail implementation methods of the simulator.

DREAM-S (Distributed REAl time Main memory database management System) is a database management system for the ATM switching system. DREAM-S has distributed system architecture to support handling of distributed data in the ATM switching system. Although, conventional main memory database management systems can guarantee a high performance, reliability of the main memory database is a critical problem since the main memory is volatile. Backup system of DREAM-S should support not only the prevention of the data loss but also the consistency of database state. In addition, the backup system has to meet the ATM switching system's requirements: real-time processing, distributed architecture, and fault-tolerance. In this paper, we propose data backup system to reduce the penalties involved in maintaining consistency and to improve data availability in switching system. Especially, we describe a concurrent log processing scheme to meet the high performance and reliability of the real-time backup system on the database update requests. And we also introduce the results of analysis to prove that the proposed technique is effective techniques.

Most real-time systems keep the whole databases in main memory for high performance. In this paper we presuppose that the system has client-server architecture with the restrictions on the memory and the time. In the light of these presupposition, we suggest an effective recovery technique for the index structures, a part of the database, and describe how it provides further performance improvement in the real-time client-server system. The basic paradigm of the proposed recovery technique is that the recovery technique of the indices should be designed differently from that of the other data part to tackle the constraints of the systems. The more client systems are appended to the system, the better performance can be guaranteed in the proposed recovery technique compared with the previous techniques.

In this paper, an efficient access method for integrating multi-scale geometric data is proposed. Previous access methods do not access multi-scale geometric data efficiently. To solve it, a few access methods for multi-scale geometric data, are known. However these methods do not support all types of multi-scale geometric data, because they support only a selection operation and a simplification operation of all map generalization operations. We propose a new method for integrating multi-scale geometric data. In the proposed method, collections of indexes in its own scale are integrated into a single index structure. By the integration, not only does the proposed method offers fast search, but also the proposed method does not introduce data redundancy. Moreover, the proposed method supports all types of multi-scale geometric data. The experimental results show that our method is an efficient method for integrating multi-scale geometric data.

An Event Service is needed for providing event delivery occurring distributed component-based applications such as multimedia communication, electronic commerce, and traffic control system. It supports asynchronous communication between multiple suppliers and consumers required by the distributed applications. However, the event service specification lacks important features for user requirements reflection as follows; reflective event filtering, user quality of service (UQoS) and component management. Thus, this paper proposes a Reflective Event Service (RES) Middleware framework for distributed component-based applications. The RES middleware based on CORBA Component Model (CCM) and includes the following components; reflective event filtering component, event monitor component, and UQoS management Component. Especially, this paper concentrates on providing suitable reflective event filtering component for UQoS service.